Seasonal variation in the association between household food insecurity and child undernutrition in Bangladesh: Mediating role of child dietary diversity

**Background:** Child undernutrition remains a major global health burden, with 149 million stunted and 45 million wasted children under 5 years in 2020. Bangladesh has made substantial progress, but stunting (28%), wasting (10%), and underweight (22%) remain high. Household food insecurity (HFI) is a known underlying factor, but its association with child undernutrition is mixed and may vary by season. Child dietary diversity (CDD) is a potential mediator. This study aimed to investigate the seasonal variation in the association between HFI and child undernutrition and the mediating role of CDD in Bangladesh.

**Methods:** Data came from the nationally representative repeated cross-sectional Food Security and Nutritional Surveillance Project (FSNSP) collected during 2012–2014. The analysis included 26,353 children aged 6–59 months. Seasons were defined based on rice harvest: Post-Aman (January–April), Monsoon (May–August, Boro harvest), and Post-Aus (September–December). HFI was measured using the HFI Access Scale and dichotomized as food secure vs. food insecure. CDD was assessed using a 7-item scale (WHO IYCF), with inadequate diversity defined as <4 food groups. Outcomes were stunting, wasting, and underweight (z-scores < -2 SD). Multivariable logistic regression was used to assess associations, adjusting for child age/sex, morbidity, maternal education/age/stature, household size/wealth, residence, and division. Causal mediation analysis (paramed command in Stata) estimated the natural indirect effect of CDD. Sample weights and complex survey design were accounted for.

**Key Results:** Overall, 36.0% of children were stunted, 11.6% wasted, and 31.9% underweight. HFI prevalence was 37.9% overall, with little seasonal variation (36.4%–39.5%). Inadequate CDD was 46.2% overall, highest in Post-Aus (51.2%). Children in food-insecure households had higher prevalence of stunting (43.5% vs. 31.5%), wasting (14.0% vs. 10.2%), and underweight (39.0% vs. 27.6%). After adjustment, HFI was associated with 12% higher odds of stunting (AOR 1.12, 95% CI 1.02–1.23), 21% higher odds of wasting (AOR 1.21, 95% CI 1.05–1.39), and 16% higher odds of underweight (AOR 1.16, 95% CI 1.04–1.30). Seasonally, the strongest associations were during Post-Aus for wasting (AOR 1.65, 95% CI 1.35–2.01) and underweight (AOR 1.21, 95% CI 1.06–1.37), and during Monsoon for underweight (AOR 1.32, 95% CI 1.08–1.62). HFI was not significantly associated with stunting in any season. Mediation analysis showed that CDD had a small but significant indirect effect on underweight overall (AOR 1.01, 95% bias-corrected CI 1.002–1.02), mediating 6.1% of the total effect. The mediation was largest during Post-Aman season, where CDD mediated 23.2% of the effect on underweight. No significant mediation was found for stunting or wasting overall, though CDD mediated 15.1% of the effect on wasting during Post-Aman.

**Clinical Implications:** The study demonstrates that HFI is a significant risk factor for child undernutrition in Bangladesh, but the association varies by season, with the strongest effects during the Post-Aus and Monsoon seasons. CDD partially mediates the HFI–underweight relationship, particularly during the Post-Aman season. These findings suggest that nutrition interventions should be season-specific, focusing on improving dietary diversity during lean periods (Post-Aman) and ensuring food security during harvest gaps. Promoting homestead food production and climate-smart agriculture could help sustain year-round access to diverse foods. The cross-sectional design limits causal inference, and the 30-day recall for HFI may not capture chronic exposure relevant to stunting. Nonetheless, the study provides nationally representative evidence to guide seasonally tailored policies and programs to reduce child undernutrition in Bangladesh.